ACCOUNTING FOR EXPERT-TO-EXPERT VARIABILITY - A POTENTIAL SOURCE OF BIAS IN PERFORMANCE ASSESSMENTS OF HIGH-LEVEL RADIOACTIVE-WASTE REPOSITORIES

Expert judgments enter several aspects of many scientific endeavors. T hey are typically employed to interpret data, predict systems' behavio ur and assess uncertainties. In particular, expert judgments are expec ted to be a relevant source of data for use in performance assessments of high-level radioactive waste repositories. In this paper we consid er the task of aggregating the judgments provided by experts with the objective of emphasizing a potential source of bias that might affect the results of the analysis. Typically, the analysts combine mathemati cal and behavioral schemes to obtain the aggregate measures desired fo r decision-making purposes. Within this approach to data aggregation, mathematical models are used as tools for sensitivity analysis and the y should account for between-expert (expert-to-expert) as well as with in-expert variability. A practical example regarding a formal expert j udgment elicitation exercise for the future climate at Yucca Mountain vicinity is presented. (C) 1997 Elsevier Science Ltd.